Robust clusterwise linear regression through trimming

Authors:
L. A. García-Escudero;A. Gordaliza;A. Mayo-Iscar;R. San Martín
Affiliations:
Departamento de Estadística e Investigación Operativa, Facultad de Ciencias, Prado la Magdalena s/n, 47005 Valladolid, Spain;Departamento de Estadística e Investigación Operativa, Facultad de Ciencias, Prado la Magdalena s/n, 47005 Valladolid, Spain;Departamento de Estadística e Investigación Operativa, Facultad de Ciencias, Prado la Magdalena s/n, 47005 Valladolid, Spain;Departamento de Estadística e Investigación Operativa, Facultad de Ciencias, Prado la Magdalena s/n, 47005 Valladolid, Spain
Venue:
Computational Statistics & Data Analysis
Year:
2010

Citing 6
Cited 6

A Classification EM algorithm for clustering and two stochastic versions

Computational Statistics & Data Analysis - Special issue on optimization techniques in statistics
A fast algorithm for the minimum covariance determinant estimator

Technometrics
Clusters, outliers, and regression: fixed point clusters

Journal of Multivariate Analysis
Principal component analysis for data containing outliers and missing elements

Computational Statistics & Data Analysis
Linear grouping using orthogonal regression

Computational Statistics & Data Analysis
Multiple model regression estimation

IEEE Transactions on Neural Networks

Editorial: Special issue on variable selection and robust procedures

Computational Statistics & Data Analysis
A class of fuzzy clusterwise regression models

Information Sciences: an International Journal
Robust fitting of mixture regression models

Computational Statistics & Data Analysis
Benchmark testing of algorithms for very robust regression: FS, LMS and LTS

Computational Statistics & Data Analysis
Robust mixture regression using the t-distribution

Computational Statistics & Data Analysis
Robust clustering around regression lines with high density regions

Advances in Data Analysis and Classification

Quantified Score

Hi-index	0.03

Visualization

Abstract

The presence of clusters in a data set is sometimes due to the existence of certain relations among the measured variables which vary depending on some hidden factors. In these cases, observations could be grouped in a natural way around linear and nonlinear structures and, thus, the problem of doing robust clustering around linear affine subspaces has recently been tackled through the minimization of a trimmed sum of orthogonal residuals. This ''orthogonal approach'' implies that there is no privileged variable playing the role of response variable or output. However, there are problems where clearly one variable is wanted to be explained in terms of the other ones and the use of vertical residuals from classical linear regression seems to be more advisable. The so-called TCLUST methodology is extended to perform robust clusterwise linear regression and a feasible algorithm for the practical implementation is proposed. The algorithm includes a ''second trimming'' step aimed to diminishing the effect of leverage points.